- 10, 12, or 16 core processors working at 4.4 GHz per core.
- As much as 256 Power11 processor cores in a single to 4 methods nodes; as much as 64 TB of 4000 MHz, DDR5 DRAM reminiscence, and 6 PCIe Gen4 x16 which 4 of them will be PCIe Gen5 x8 and two PCIe Gen5 x8.
- 19-inch PCIe Gen4 4U I/O growth drawer and PCIe fan-out modules, supporting a most of 192 PCIe slots and 4 I/O growth drawers per node.
- PCIe Gen1, Gen2, Gen3, Gen4, and Gen5 adapter playing cards supported within the system node, and PCIe Gen1, Gen2, Gen3, and Gen4 adapter playing cards supported in I/O growth drawer.
- Dynamic LPAR assist for adjusting workload placement of processor and reminiscence assets.
- Lively Reminiscence Enlargement (AME) that’s optimized onto the processor chip.
- Energy Enterprise Swimming pools that assist unsurpassed enterprise flexibility for workload balancing and system upkeep.
The brand new midrange server, the 4U rack mounted IBM Energy E1150 helps 32 to 120 processor cores, 256 GB -16 TB high-performance as much as 4000 MHz DDR4 OMI reminiscence and Elastic, Cell and Shared Utility Capability choices that means the Energy E1150 system lets shoppers deploy pay-for-use consumption of processor, reminiscence, and supported working methods, by the day, throughout a group of Energy E1150 methods, IBM said. The S1122is a2U rack-mounted, entry-level Power11 server can assist both two Entry Single-Chip Module (eSCM) or Twin Chip Mode (DCM) processors per server relying on buyer energy wants. Every new Power11 processor single chip module (SCM) accommodates two reminiscence controllers. 4 10-core 3.90 – 4.20 GHz (max), 4 12-core 3.90 – 4.40 GHz (max), or 4 16-core 3.80 – 4.30 GHz (max) are utilized in every system node, offering 40 cores to a 160-core system, 48 cores to a 196-core system, or 64 cores to a 256-core system, in line with IBM.
“The brand new Power11 platform has been particularly designed to assist organizations harness alternatives within the period of AI and hybrid, multicloud operations,” wrote Drake. “First, Power11 addresses each efficiency and compliance necessities of AI, together with acceleration capabilities and the flexibility to entry, remodel, and handle enterprise knowledge at scale, and assist the seamless integration of generative AI (genAI) capabilities with mission-critical enterprise processes,” Drake wrote.
Second, the platform helps new automation capabilities, bringing a variety of advantages to Energy platform clients that span administration, reliability, safety, and sustainability, Drake wrote. “Lastly, IBM’s Power11 is designed to assist a distributed and versatile hybrid infrastructure with business-driven workload placement throughout on premises, public cloud, or personal cloud based mostly on efficiency, value, and compliance wants.”
The entire servers embrace an AI-based Matrix Math Accelerator that helps to carry out in-core AI inferencing and machine studying the place knowledge resides, IBM said. IBM says there will be zero deliberate downtime for system upkeep. As an illustration, options equivalent to autonomous patching, automated workload motion, and deliberate system upkeep occasions can happen with out ever taking essential purposes offline, in line with Bargav Balakrishnan, vice chairman of product administration with IBM. This characteristic can free IT professionals from spending time planning, testing, and executing upgrades to their methods, Balakrishnan mentioned. The Energy servers assist IBM’s Energy Cyber Vault that utilized AI/ML and the NIST cybersecurity framework to assist determine, defend, detect, and mechanically reply to cyber threats immediately, Balakrishnan mentioned.
“Cyber Vault supplies safety in opposition to cyberattacks equivalent to knowledge corruption and encryption with proactive immutable snapshots which can be mechanically captured, saved, and examined on a custom-defined schedule. Power11 additionally used NIST-approved built-in quantum-safe cryptography designed to assist defend methods from harvest-now, decrypt later assaults in addition to firmware integrity assaults,” Balakrishnan mentioned.
Different shared server options embrace:
